Search Results (264)

Cementitious sand and gravel (CSG) is commonly used in construction engineering; however, its mix proportion design is complex, and traditional physical experiments face limitations such as long cycles, high costs, and susceptibility to external factors when obtaining high-quality sample data. In this study, a foundational dataset was first acquired through physical experiments: 100 sets of CSG specimens with different mix proportions (cement content 40, 50, 60, 70 kg/m³; water-to-binder ratio 1.0, 1.2, 1.4; sand ratio 0.1, 0.2, 0.3, 0.4; fly ash content 20, 30, 40, 50 kg/m³) were prepared. After 28 days of standard curing, compressive strength and splitting tensile strength tests were conducted using a WAW-1000 electro-hydraulic servo universal testing machine, yielding 100 sets of real mechanical property data. The coefficients of variation for all test groups were below 10%, confirming the reliability and repeatability of the experimental data. On this basis, a data augmentation method based on Conditional Tabular Generative Adversarial Networks (CTGAN) is proposed. Through adversarial training between the generator and the discriminator, the model learns the multi-dimensional distribution characteristics of the original CSG data and generates 100 synthetic samples, which are then merged with the original data to expand the dataset to 200 samples. The quality of the synthetic data is evaluated using Wasserstein distance and correlation matrix heatmaps. Furthermore, a hybrid XGBoost–LSTM prediction model is proposed—XGBoost is used for feature construction to capture nonlinear interactions among mix proportion variables, and the constructed features are then fed into an LSTM network for sequential learning and regression prediction. The results show that the CTGAN-generated data are highly consistent with the original data in terms of kernel density distributions and variable correlations, with Wasserstein distance significantly superior to four comparative methods: Bootstrap, SMOTE, GaussianCopula, and TVAE. After augmentation, the XGBoost–LSTM model achieves a coefficient of determination (R²) of 0.9897 for compressive strength prediction (vs. 0.9793 before augmentation) and 0.9801 for splitting tensile strength (vs. 0.9882 before augmentation, a slight decrease). The mean absolute percentage errors (MAPE) are 4.49% and 4.11%, and the root mean square errors (RMSE) are 0.201 and 0.049, respectively; both error metrics are reduced compared with those before augmentation. Compared with baseline models including XGBoost, LSTM, Random Forest (RF), and Support Vector Regression (SVR), the XGBoost–LSTM model exhibits the best performance across all evaluation metrics, and Wilcoxon signed-rank tests confirm that the performance differences are statistically significant (p < 0.05). The proposed method of CTGAN-based data augmentation combined with the XGBoost-LSTM hybrid model provides an effective solution to the problem of insufficient CSG sample data and offers a reference for data enhancement and performance prediction of other small-sample materials. Full article

Background: Eastern Tongren City, Guizhou Province, China, possesses abundant wild germplasm resources of Camellia oleifera Abel.; however, there is a lack of systematic evaluation, and the promotion of superior varieties is insufficient. This study aimed to evaluate 21 trait indices of 189 wild C. oleifera accessions from four regions in the Tongren area to clarify their variation characteristics, assess the effects of altitude on trait expression, and identify candidate germplasms with outstanding comprehensive performance. Methods: A total of 21 traits spanning fruit morphology, oil content, fatty acid composition, and bioactive components (tocopherols, squalene, and polyphenols) were measured. Principal Component Analysis (PCA) was employed to construct a comprehensive evaluation score (Z_n) for quantitative ranking and screening. Trait differences between a low-altitude group (400–800 m, n = 139) and a high-altitude group (800–1200 m, n = 50) were compared using Welch’s t-test. Results: The germplasms exhibited abundant phenotypic variation, with coefficients of variation (CV) ranging from 5.29% (total unsaturated fatty acids) to 124.42% (beta + gamma-tocopherol). Bioactive components showed the highest variability, while fatty acid composition was relatively stable. Altitude had a significant effect on six of the 21 traits. Seed oil content and kernel oil content were significantly higher in the high-altitude group, with mean differences of 5.92 and 5.74 percentage points, respectively (both p < 0.001). However, oleic acid, total unsaturated fatty acids, fruit morphological traits, and most bioactive components showed no significant altitudinal differences (p > 0.05). The first five principal components explained 65.0% of the total variance. CL40 achieved the highest comprehensive score (Z_n = 4.15), followed by MJX2 (Z_n = 2.56). Among the top 10 individuals, eight were from the low-altitude group. Conclusions: This study revealed rich phenotypic variations and distinct altitudinal effects among wild C. oleifera germplasms in eastern Guizhou. The superior germplasms identified (such as CL40 and MJX2) can serve as candidate materials for locally adapted variety improvement. This study was primarily based on single-season phenotypic data, and the genetic stability of the selected germplasms should be validated through clonal trials and molecular marker analysis in future research. Full article

To address the low accuracy and poor reliability of short-term photovoltaic (PV) power forecasting under complex weather conditions, this study proposes a multi-strategy synergistic optimization framework for point-interval prediction. The methodology integrates similar day classification (SDC) via RDP-DTW-DBA-K-means, a hybrid BiTCN-MAOBiGRU-AM model with a mutation-aware mechanism, and Dream Optimization Algorithm (DOA) for global hyperparameter tuning of six key parameters. For interval prediction, an adaptive bandwidth kernel density estimation (ABKDE) dynamically adjusts bandwidth based on local error density and weather scenarios. Experiments using data from a Guangxi PV station demonstrate that the synergistic model reduces RMSE by 29.58% (cloudy) and 32.37% (overcast/rainy) versus the baseline, and cuts RMSE by 16.2–19.0% under abrupt weather events and 22.4–40.2% under non-ideal input data. At the 95% confidence level, ABKDE improves prediction interval coverage probability by 3.9–5.4 percentage points and reduces normalized average width by 20.8–23.6% compared to conventional KDE. The proposed framework significantly enhances prediction accuracy, robustness, and generalization, offering a reliable solution for PV power forecasting in highly variable meteorological scenarios. Full article

Precise prediction of voltage degradation is critical for the prognostics and health management of proton exchange membrane fuel cells (PEMFCs). The performance degradation of PEMFCs is governed by a complex coupling of multiple physicochemical mechanisms, including catalyst layer and proton exchange membrane degradation. Crucially, these internal degradation processes evolve across highly heterogeneous time scales, ranging from transient high-frequency fluctuations to low-frequency and long-term irreversible performance fade. Conventional predictive models, which typically rely on single-scale architectures or fixed receptive fields, are inherently ill-equipped to simultaneously decouple and capture these cross-scale temporal dynamics. To tackle this challenge, this paper innovatively proposes a multi-scale deep learning framework that integrates a multi-scale degradation trend perception module, a long short-term memory (LSTM)-based encoder–decoder architecture, and a multi-head attention mechanism. One-dimensional convolutional layers with different kernel sizes are employed to simultaneously extract local temporal features at multiple granularities, followed by the LSTM encoder–decoder to model long-range temporal dependencies, while the cross-attention mechanism dynamically allocates attention across the encoded context at each autoregressive decoding step. Experimental outcomes indicate that the proposed model realizes excellent predictive accuracy across five evaluation indices in comparison with standard baselines. In particular, the mean absolute percentage error (MAPE) reaches 1.6696%, and the maximum absolute percentage error (Max-APE) is strictly bounded within 5%, substantiating the reliability of the proposed framework for high precision and long-horizon health prognostics for PEMFCs. Full article

Rice-milling quality is conventionally inspected post-process, leaving operators unable to correct breakage as it occurs. We present and quantitatively validate a cloud-mediated closed-loop architecture that couples a YOLO26 machine-vision pipeline to Arduino-based actuator control on a laboratory rice mill. The image-acquisition node uploads frames to a cloud repository; an inference and analysis node retrieves them, runs YOLO26 detection with a hybrid post-process classifier to estimate the broken-rice fraction, and issues a command to an Arduino microcontroller that drives PWM-modulated motor and vibrator actuators. The detector achieved a mean Average Precision of 0.951 (peak precision 0.99, peak recall 0.98) on a held-out test set of 100 images. In a matched comparison against an open-loop baseline ($n=196,000$ kernels, broken fraction 21.04%), closed-loop operation ($n=114,000$ kernels) reduced the broken fraction to 6.19%, an absolute improvement of 14.85 percentage points (two-proportion z-test: $z=112.8$, $p<0.001$, 95% CI for the absolute reduction: 14.62–15.08 pp). Dynamic analysis identified a near-linear plant gain of 1.0–1.5% breakage per 1% PWM, providing the empirical basis for future formal PID and Model Predictive Control synthesis. The principal empirical contribution is a quantitative characterisation of the PWM-to-breakage transfer relationship of a rice-milling actuator under deep-learning-derived quality feedback, together with a matched open-loop/closed-loop demonstration that this feedback loop moves the laboratory prototype from non-compliant to Grade A-equivalent quality at constant throughput. The lab-scale prototype is not yet industrial; a roadmap to pilot-scale deployment is outlined. Full article

Knowledge distillation transfers information from a high-capacity teacher to a smaller student, but its behavior across regression, classification, and heterogeneous tabular model families remains insufficiently understood. This paper presents a comparative simulation study of distillation in structured-data settings, where neural networks, tree ensembles, kernel methods, nearest-neighbor methods, and linear models are plausible competitors. Synthetic datasets with known ground truth and controlled noise are used to evaluate a three-model teacher ensemble distilled into multilayer perceptrons, tree ensembles, decision trees, kernel methods, instance-based methods, and linear models. Distillation weights and temperatures are selected on validation data, and test results are reported with confidence intervals, paired tests, and multiplicity-control checks. In regression, distillation acts as target smoothing and denoising: the teacher reduces label noise by 17.4%, 13 of 14 students improve numerically, and the MSE reduction is 7.1%. After multiplicity control, robust gains concentrate among MLPs. In classification, gains are smaller: 12 of 15 students improve, with an average accuracy gain of 0.5 percentage points, and soft-label distillation improves calibration for MLPs. Additional noise sweeps, teacher ablations, real-data checks, baseline comparisons, temperature analysis, and calibration results show that distillation helps most when the student has sufficient capacity, and the teacher provides a cleaner or more informative target. An extended set of confirmatory analyses—including teacher-quality controls with a shuffled negative control, ensemble-based uncertainty proxies, harder data conditions, distribution-shift robustness, a capacity ladder, computational-cost analysis, additional baselines and calibration metrics, an $\alpha$–T factorial analysis, quantitative measures of soft-probability structure, and comparisons with modern tabular models—shows that useful transfer requires teacher predictions that remain conditionally aligned with the inputs. The classification gains are small in absolute accuracy but stable across seeds and are accompanied by improvements in selected calibration measures. Full article

Vibration-based mechanical fault diagnosis has become a critical research area, mostly driven by the need to improve equipment reliability and reduce unplanned downtime in industrial settings. Time–Frequency Networks (TFNs) have shown strong potential here, combining interpretable time–frequency transformations with deep learning classifiers in a single framework. This work reproduces the original TFN model from the recent literature and evaluates its noise robustness under additive Gaussian noise (10 dB, 0 dB, $-5$ dB SNR) and impulsive noise at the same levels, across five architectures: Backbone CNN, Random CNN, TFN-Chirplet, TFN-Morlet, and a squeeze-and-excitation attention CNN baseline. The evaluation protocol corrects two methodological issues identified during peer review of an earlier version of this work—window-level data leakage between train and test splits, and selection of the best-performing training epoch rather than a fixed final-epoch result—both of which are shown to materially affect reported outcomes. Under the corrected protocol, TFN-Morlet remains the most noise-robust architecture, with only a 19.09% accuracy drop from clean to $-5$ dB AWGN, approximately 15.5 percentage points better than Backbone CNN under the same conditions; an architectural anomaly reported in the earlier version of this study, in which mild noise appeared to improve an unconstrained CNN’s accuracy, was not reproduced under the corrected protocol and is shown to be an artifact of the original methodological issues. Per-class analysis and multi-model confusion matrices further reveal that misclassifications under severe noise are dominated by confusion between the same defect severity at different fault locations, rather than between different severities at the same location as previously reported. These results indicate that time–frequency-aware convolutional kernels improve both classification accuracy and noise resistance under rigorous, leakage-free evaluation, and that this robustness is not replicated by a generic attention mechanism alone. Full article

Greenhouse tomato detection faces critical challenges due to dense fruit occlusion, background interference, and the stringent computational constraints inherent to edge-deployed harvesting robots. Resolving these bottlenecks requires efficient architectures. We propose LGD-YOLO as an asymmetric lightweight network adapted from YOLOv10n specifically for edge-based tomato maturity detection. The architecture integrates a C2f-GMKSF module utilizing grouped multi-kernel convolutions to extract multi-scale textures with limited computational overhead. Precise feature alignment under occluded conditions is subsequently achieved through a Dy-HSFPN structure, accompanied by a C2f-CFCGLU module that expands the receptive field while preserving linear complexity. Furthermore, replacing the traditional detection head with a partial convolution head reduces memory access costs. A Focaler-Wise-SIoU loss function is utilized to stabilize bounding box regression against the lightweight penalty without introducing inference latency. Performance evaluations on a custom three-class dataset with a 180-image test set yield an 88.0% mAP@50. Relative to the baseline model, LGD-YOLO improves detection accuracy by 0.6 percentage points while shrinking the parameter volume by 37.6% to 1.41 M and lowering computational demand by 41.5% to 3.8 GFLOPs. Hardware deployment on an NVIDIA Jetson AGX Orin achieves a sustained processing speed of 40.6 FPS, while the weight file is 2.99 MB, supporting its feasibility for real-time agricultural robotics. Full article

Marine oil spills pose serious threats to coastal ecosystems and maritime activities, and synthetic aperture radar (SAR) has become an important tool for all-weather marine monitoring. However, SAR oil spill detection remains challenging because oil spills usually appear as weak dark anomalies with blurred boundaries, elongated or fragmented shapes, and strong interference from lookalike phenomena such as low-wind areas and internal waves. To address these issues, we propose RGCNet, a lightweight radiometric–geometric–contextual detection framework based on YOLOv11n. Firstly, the H_SPDRFF module is incorporated into the backbone to enhance weak radiometric responses through constrained feature amplification, thereby reducing missed detections caused by low-contrast oil slicks. Secondly, the C3k2_GSR module is designed in the neck to strengthen anisotropic geometric refinement and preserve the continuity of elongated and fragmented oil spill regions during multi-scale feature fusion. Finally, a SAR-adapted large selective kernel block (LSKBlock) is embedded in the high-level backbone to improve contextual discrimination between true oil spills and lookalike dark formations. Experiments on DeepSAR show that RGCNet increases mAP@0.5 and mAP@0.5:0.95 by 3.6 and 3.0 percentage points over the YOLOv11n baseline, respectively. Cross-dataset evaluation on SAR-Oil-Spill demonstrates a 3.9-point mAP@0.5 gain, indicating strong transferability. Furthermore, with a compact model size of 2.67 M parameters and 6.4 G FLOPs, RGCNet achieves an inference speed of 162.5 FPS on an RTX A4000 GPU, demonstrating its efficiency and potential for real-time maritime surveillance. Nevertheless, the current bounding-box formulation cannot precisely delineate irregular oil-spill boundaries. Future work will therefore investigate fine-grained segmentation and cross-sensor adaptation. Full article

To address the challenges in maize leaf disease detection, including large variation in lesion scales, weak texture of small targets, strong background interference, limited recall ability for blurred lesions, and computational redundancy of conventional detection heads, this paper proposes a lightweight detection algorithm based on an improved YOLOv11n. First, a multi-scale global context kernel attention module is designed, which employs GCKA-bottleneck with large-kernel attention and residual connections to enhance the deep semantic representation of multi-scale lesions. Second, a GSConv-enhanced coordinate multi-receptive attention module is constructed, which combines coordinate position awareness and multi-scale depthwise convolution. Finally, a regression-enhanced depthwise-separable decoupled detection head is proposed to decouple classification and regression tasks, and introduces depthwise separable convolution and a distributed bounding box regression. On a public dataset containing four classes, the improved model achieves an mAP@0.5 of 85.36%, a recall of 83.32%, and a precision of 86.91%, which are 3.46, 27 2.94, and 2.38 percentage points higher than those of the original YOLOv11n, respectively. Meanwhile, GFLOPs and parameter count are reduced by 27.0% and 12.4%, respectively. The proposed algorithm strikes a favorable balance between accuracy, real-time performance, and lightweight design, providing a feasible technical support for field deployment in intelligent agricultural disease monitoring systems. Full article

Accurate detection and counting of open cotton bolls from unmanned aerial vehicle (UAV) RGB imagery are essential for organ-level cotton phenotyping and field monitoring, but remain challenging in high-density cotton fields because open bolls are small, densely distributed, partially occluded, and visually similar to plastic mulch, branches, senescent leaves, shadows, and drip-irrigation belts. To address these challenges, this study proposes OpenBoll-YOLO, a lightweight small-object detector designed for open cotton boll detection and counting in UAV nadir-view images. A UAV RGB dataset was collected at the boll-opening stage in Alar, Xinjiang, China, covering two cotton varieties and two acquisition dates. Based on YOLOv11s, OpenBoll-YOLO integrates three task-oriented components: a multi-kernel small-object enhancement pyramid (MSOEP) to preserve shallow spatial details and strengthen multi-scale feature fusion, a cross-stage partial block with a dynamic mixing layer (C2DML) to improve local structural discrimination under complex backgrounds, and a lightweight mixed aggregation network (LMANet) to enhance contextual representation with reduced model complexity. On the independent test set, OpenBoll-YOLO achieved 86.7% precision, 84.5% recall, 85.6% F1-score, 92.9% mAP@0.5, 81.5% mAP@0.75, and 71.7% mAP@0.5:0.95, with only 3.1 M parameters and an inference speed of 86 frames s⁻¹. Compared with YOLOv11s, it improved mAP@0.5 and mAP@0.5:0.95 by 2.4 and 5.9 percentage points, respectively, while reducing the parameter count by 67.0%. Counting evaluation further showed that OpenBoll-YOLO reduced the mean absolute error from 11.98 to 10.12 bolls image⁻¹ and increased R² from 0.87 to 0.91. These results demonstrate that OpenBoll-YOLO provides an accurate and lightweight solution for dense open cotton boll detection and counting in high-density field conditions. Full article

Potato leaf diseases directly reduce yield and quality, and accurate field detection is important for precision plant protection. However, potato disease lesions are often weak in deep semantic representation, easily disturbed by complex field backgrounds, and variable in multi-scale lesion texture. To address these challenges, this study proposes an improved YOLO-based potato leaf disease detection model. The proposed model enhances the detector through three task-oriented modules. Deep Symptom Enhancement is used to strengthen deep disease feature extraction. Lesion Selection Attention based on large separable kernel attention improves the spatial selection of lesion regions. Multi-Scale Refinement Adapter uses a Mona-based C2PSA structure with two stacked Mona adapters to refine multi-scale texture and lesion-boundary information. Experiments were conducted on a potato leaf disease image dataset using mAP50, average recall (AR), parameters, GFLOPs, and FPS as evaluation metrics. The baseline YOLO26s achieved 81.31% mAP50 and 77.85% AR. The proposed SLR-YOLO model achieved 88.92% mAP50 and 83.51% AR, improving mAP50 and AR by 7.61 and 5.66 percentage points, respectively, while maintaining 118.6 FPS. The results show that the proposed framework improves detection accuracy for potato leaf disease images while retaining practical real-time performance. Full article

The marine capture fisheries sector of the Philippines employs approximately 2.3 million Filipinos, yet recent declines (including a 15.3% drop in Q1 2026 production relative to Q1 2025) underscore the need for forecasting systems resolved at the regional and sectoral level. Existing Philippine approaches rely on univariate classical time-series methods and seldom integrate multivariate oceanographic predictors. This study addresses three questions: (RQ1) How do nine candidate machine learning algorithms compare in forecasting regional fish production from environmental predictors? (RQ2) Which environmental predictors most strongly drive model output, as quantified by explainable AI (XAI) SHAP-based feature attribution? (RQ3) To what extent do model performance and predictor importance vary across regions? Across 32 region–sector panels spanning 2002–2025, kernel and neural network models were selected as the best-performing architecture in 26 of 32 panels (81.3%), achieving a mean composite score 12.7% higher than tree-based ensembles, a gap attributable to extrapolation along trending physical predictors. Feature attribution identified the partial pressure of CO₂ as the leading driver in both sectors, exceeding the second-ranked variable by factors of 2.5 (commercial) and 3.4 (marine municipal). Regional heterogeneity in retained predictors, winning algorithms, and SHAP attribution rankings supports region-specific forecasting as a necessary design choice. Mean absolute percentage error of 22–25% and directional accuracy of 0.62–0.66 indicate operational utility for early-warning applications, establishing a basis for evidence-driven priority-setting in Philippine fisheries governance. Full article

Accurate grade detection in graphite ore, which is a strategic and critical mineral resource, plays an important role in improving beneficiation efficiency and overall resource utilization. However, the scarcity of high-grade samples limits the performance of deep learning models in grade identification tasks. This limitation makes it difficult for models to learn stable and representative features. This paper proposes an enhanced CycleGAN-based image augmentation framework designed for graphite ore imagery. The method works within an unpaired image translation architecture. It introduces a distributed reflectance consistency loss. This loss encodes the graphite ore’s typical low reflectance and high optical contrast as explicit statistical constraints. The design enforces consistency in both the intensity distribution and the textural structure of the generated images. The model further integrates a convolutional block attention module into the generator. This module helps refine feature representation under a physics-inspired heuristic. The study constructs augmented training sets using the proposed method. It then evaluates these datasets with a downstream grade classification model. Experimental results show clear improvements. The method reduces Fréchet Inception Distance by 21.9% and Kernel Inception Distance by 39.4%. It also improves peak signal-to-noise ratio by 3.3% and structural similarity index measure by 2.6% compared with the baseline CycleGAN. The classification accuracy in the grade identification task increases by about 2.3 percentage points. These results show that the proposed method improves both the perceptual quality and the statistical consistency of synthetic graphite ore images. It also helps reduce the performance drop caused by limited training data in few-shot learning conditions. Full article

To investigate the adoption behavior and decision-making mechanism of mechanized ear harvesting for seed corn, this study analyzed 786 household survey data from the Hexi Corridor using binary logistic regression, marginal effect analysis, interaction effect tests, and mediation models. The results revealed that the overall adoption rate of mechanized ear harvesting stands at 31.8%, with significant variations across regions and farm sizes. Adoption was found to follow a sequential decision chain: ‘technical feasibility → economic feasibility → comparative benefit assessment.’ Parental line lodging resistance (OR = 3.48) and field contiguity (OR = 3.01) were shown to positively influence adoption indirectly, mediated by perceived harvest loss reduction and perceived machinery efficiency enhancement. A 1% increase in kernel breakage rate is associated with a 2.7 percentage-point reduction in adoption probability. Enterprise-organized mechanical services were identified as the strongest adoption driver (OR = 6.19) and were found to function as a ‘scale equalizer’, significantly reducing the adoption advantage of larger farms (interaction coefficient B = −0.043, p = 0.024). Additionally, a pronounced scale-threshold effect is identified: adoption rates are observed to rise sharply beyond 3.33 hm², contradicting linear scale-adoption assumptions. These findings highlight the critical roles of inclusive enterprise services, a minimum efficient scale, and the sequential decision process. Coordinated innovation across breeding, equipment engineering, and extension systems is required for sustainable mechanization. Full article